C1632 suppresses the migration and proliferation of non‐small‐cell lung cancer cells involving LIN28 and FGFR1 pathway

Abstract Chemoresistance and migration represent major obstacles in the therapy of non‐small‐cell lung cancer (NSCLC), which accounts for approximately 85% of lung cancer patients in clinic. In the present study, we report that the compound C1632 is preferentially distributed in the lung after oral administration in vivo with high bioavailability and limited inhibitory effects on CYP450 isoenzymes. We found that C1632 could simultaneously inhibit the expression of LIN28 and block FGFR1 signalling transduction in NSCLC A549 and A549R cells, resulting in significant decreases in the phosphorylation of focal adhesion kinase and the expression of matrix metalloproteinase‐9. Consequently, C1632 effectively inhibited the migration and invasion of A549 and A549R cells. Meanwhile, C1632 significantly suppressed the cell viability and the colony formation of A549 and A549R cells by inhibiting DNA replication and inducing G0/G1 cell cycle arrest. Interestingly, compared with A549 cells, C1632 possesses the same or even better anti‐migration and anti‐proliferation effects on A549R cells, regardless of drug resistance. In addition, C1632 also displayed the capacity to inhibit the growth of A549R xenograft tumours in mice. Altogether, these findings reveal the potential of C1632 as a promising anti‐NSCLC agent, especially for chemotherapy‐resistant NSCLC treatment.


| INTRODUC TI ON
Lung cancer is one of the most common malignant tumours and is responsible for 25% of cancer-related deaths each year. 1,2 Approximately, 85% of lung cancer patients have been clinical diagnosed as non-small cell lung cancer (NSCLC); thus, the treatment of NSCLC has been an urgent health issue worldwide. 3 Progress in this area has been substantial and promising over the past 20 years with the advent of various targeted therapies 4 and immunotherapy 5 in some advanced NSCLC patients. 6 For instance, the use of small molecule tyrosine kinase inhibitors, such as EGFR tyrosine kinase inhibitor, 7-11 ALK inhibitors 12,13 and ROS1 inhibitors, 14 has achieved unprecedented survival benefits in some selected patients.
However, small molecule tyrosine kinase inhibitors could only be used for a small minority of NSCLC patients with gene alterations. 15 Consequently, the overall cure and survival rates of NSCLC remain low. 1,16 Thus, continued research into new small molecule inhibitors that significantly suppress NSCLC cell motility and invasiveness as well as proliferation is desired.
LIN28, which is an RNA-binding protein consisting of LIN28A and LIN28B, 17 is an important regulator of miRNAs and mRNAs. 18,19 LIN28 regulates not only the translation of mRNAs that play a key role in cell growth and metabolism but also the biogenesis of miRNAs. 20,21 Recently, studies have found that LIN28 levels are increased in clinical NSCLC tissues, and this is correlated with the increased ability of cell migration and proliferation, [22][23][24] and drug resistant of human lung cancer cells. 25 Thus, LIN28 is an appealing therapy target for small molecule drugs in the NSCLC treatment field.
Fibroblast growth factor receptor 1 (FGFR1), an oncogenic receptor tyrosine kinase, plays fundamental roles in stimulating cell proliferation and migration. 26 Importantly, FGFR1 expression is found to be substantially increased in clinical NSCLC tissues compared with the adjacent peritumoural tissues. 27,28 Meanwhile, several FGFR1 inhibitors are under early-phase clinical trials of NSCLC treatment. [29][30][31] For instance, AZD4547 (AstraZeneca, NCT01213160) and BGJ398 (Novartis; NCT01004224; NCT01697605) possessed potential benefits in advanced NSCLC patients. [32][33][34] All these studies suggested that FGFR1 inhibition could serve as an important therapeutic target in NSCLC. 27 Unfortunately, FGFR1-targeted therapies would result in drug resistance which finally led to the failure of targeted therapies. 1,26,28 Thus, it is reasonably speculated that multitargeted small molecule inhibitors that inhibit FGFR1 along with other functional proteins, such as LIN28, may exert the strongest effects in the clinical treatment of NSCLC.
The anxiolytic compound C1632, 35,36 has been demonstrated to have LIN28 inhibitory effect in our previous work. 37 In the present study, we discovered that C1632 mainly accumulates in lungs of mice after oral administration with high bioavailability.
Furthermore, C1632 suppressed the expression of LIN28 as well as the phosphorylation of FGFR1 in a dose-dependent manner in NSCLC A549 and A549R cells. We also investigated the possibility that C1632 acts as a small molecule anticancer drug targeting both LIN28 and FGFR1 in the treatment of NSCLC, especially for cisplatin-resistant NSCLC.

| Cell adhesion assay
The cell adhesion assay was performed as indicated previously. 38 Briefly, the 96-well plate was coated with human fibronectin (Millipore), and then cells were seeded into the 96-well plate and cultured for another 1 h at 37°C in a 5% CO 2 incubator. Finally, rinsing the plate before fixed with 10% formalin and stained with crystal violet. As followed, wash the plate with ddH 2 O and add 100 μl acetic acid (33%) into the plate to dissolve the crystal violet. The absorbance at 560 nm was detected by Synergy H1 Multi-Mode Reader (BioTek). Relative number of cells attaching to extracellular matrix was evaluated using the following equation: mean OD of treated cells/mean OD of control cells.

| Transwell assay
Transwell assay was carried out according to the manufacturer's instructions provided by Transwell Kit (Corning Costar). Briefly, cells were pretreated with indicated concentrations of C1632 for 5 days, then harvested and re-seeded into insert (transwell permeable support) containing 100 μl serum-free DMEM medium. The insert was placed into 24-well plate containing 600 μl of DMEM medium extra added with 10% FBS. 24 h later, cells on the upper surface of the insert were removed with cotton-tipped swabs. And cells on backside surface of the insert were fixed with 10% formalin, then stained with crystal violet. The insert was washed three times with ddH 2 O before it is subjected to Nikon Ti microscope observation. Additionally, these inserts were dissolved in 500 μl acetic acid (33%) separately, and the absorbance at 560 nm was detected by the spectrophotometer (DTX880, Beckman Coulter).

| Scratch-wound assay
The cells scratch-wound assay was performed as previous reported. 38 The cells were seeded in a 6-well plate and then cultured in DMEM medium containing indicated concentration of C1632 or 0.01% DMSO for 5 days. A denuded area was created across the diameter of dish by a yellow tip as the cell density up to 95%. Then cells were maintained in a serum-free medium throughout the test.
Phase-contrast images were taken at the indicate time by Nikon Ti microscope and analysed with Axiovision Rel.4.8 software.

| Immunofluorescence assay
Performing the immunofluorescence (IF) assay as previously described. 39 Briefly, cells were seeded to slide, fixed with 4% paraformaldehyde, and permeabilized using 0.5% Triton X-100. Next, cells were incubated with primary antibodies against FAK (CST), followed by incubating with secondary antibodies (DyLight 488-conjugated anti-mouse). DAPI was used to stain the nuclear. The fluorescence image was captured by using Nikon Ti microscope and the quantitative analysis was carried out by ImageJ.

| Edu staining assay
Edu staining assay was carried out according to Edu staining Kit (Beyotime). First, A549 or A549R cells were seeded in 6-well plates and cultured in RPMI medium 1640 containing 10% FBS, then treated with C1632 (15, 30,

| Colony cloning assay
First, cells were treated with indicated concentrations of C1632 or 0.01% DMSO for 5 days. Then these cells were separated into single cells that were directly used for cloning. During the process of cloning, C1632-treated A549 or A549R cells were still maintained in DMEM plus 10% FBS medium containing the indicated concentration of C1632 (15,30, and 60 mg/L), while the control group was cultured with 0.01% DMSO. Both culture media were changed for every 2 days until 10 days. The number of forming colonies in C1632 or 0.01% DMSO-treated groups was counted and the images were taken.

| Cell cycle distribution analysis
Cells were cultured in the absence or presence of 15, 30, and 60 mg/L of C1632 or 0.01% DMSO for 5 days, trypsinized, washed, and stained with propidium iodide before cell cycle distribution was assessed on a flow cytometer (BD FACSCalibur, BD Biosciences).

| Annexin V/PI apoptosis assay
Cells were seeded in 6 cm 2 dish with a density of 3.0 × 10 5 cells per dish and treated with C1632 with final concentration of 15, 30, and 60 mg/L or 0.01% DMSO for 5 days. Then the assay was performed following the protocol provided by the Annexin V/PI Apoptosis Kit (Sigma) and was assessed on a flow cytometer (BD FACSCalibur, BD Biosciences).

| Senescence-associated β-galactosidase (SAβ gal) activity assay
The assay was set up following the β-galactosidase (SAβ gal) staining Kit, which was obtained from Sigma. In brief, cells were washed once with PBS and fixed with stationary liquid provided in the kit at room temperature for 15 min. Next, the cells were incubated overnight at 37°C in dark with the 1 ml working solution containing 0.05 mg/ml 5-bromo-4-chloro-3-indolyl-b-d-galactopyranoside (Xgal) and observed under a normal light microscope (Nikon).

| RNA extraction and real-time quantitative RT-PCR
Total RNA was extracted using RNAiso Plus (Takara) according to the manufacturer's instructions. cDNA was synthesized using PrimeScript II first-strand cDNA Synthesis Kit (Takara), followed by amplification with RealStar Power SYBR Mixture (GenStar). miR-NAs were quantified using the stem-loop method, and U6 snRNA was adopted as internal control. Specific primer sequences were designed as previously reported. 40

| Animal studies
The  As followed, the samples were centrifuged at 14,954 g for 10 min.

| Development of UHPLC-MS/MS method for determining C1632
Add acetonitrile (400 µl) and IS (20 µl) into collected plasma samples (100 µl). Thereafter, the samples were vortexed for 2 min, followed by centrifugation at 14,954 g for 10 min. Remove the supernatants to 1.5 ml tube and the sample is ready for detection by established UHPLC-MS/MS assay. The injection volume is 6 µl. The pharmacokinetic parameters were determined using DAS software (Version 3.0).

| Tissue distribution study
Twenty-four mice were randomly divided into four groups (six mice for each group, one group for each time point) and received 20 mg/kg (i.v.) of C1632 by oral administration. The mice were euthanized by decapitation at 0 (blank group), 0.25, 2 and 6 h after C1632 was given. Tissues were collected and washed with normal saline, then homogenized and subjected to sample preparation. Subsequently, the concentration of C1632 was determined by UHPLC-MS/MS.

| Microsomal incubation assay
The microsomal incubation assay 41 was performed at 37°C in a

| Statistical analysis
Statistical analyses were performed with GraphPad Prism 5 using a two-tailed Student t-test or two-way ANOVA with the p values. Values are presented as mean ± standard deviation (SD). p values < 0.05 were considered statistically significant (*p < 0.05; **p < 0.01; ***p < 0.001).

| C1632 is preferentially distributed in the lung after oral administration in vivo with high bioavailability and limited inhibitory effects on CYP450 isoenzymes
The concentration-time curves of C1632 ( Figure 1A) in mouse plasma, heart, liver, spleen, lung, kidney, and brain after oral administration (20 mg/kg) are shown in Figure S1. The tissue distribution results indicated that C1632 diffuses rapidly and widely into major organs, with a peak at 0.25 h ( Figure S1). The level of C1632 was highest in the lung and liver, followed by the kidney, heart, and spleen ( Figure 1B).
As expected, the highest accumulation took place in the liver, where the majority of medicine metabolism takes place. The level of C1632 in the brain remained low, suggesting that C1632 may be effectively prevented from crossing the blood-brain barrier. The fact that the accumulation in the lung was equal to that in the liver indicates that C1632 has potential application in lung cancer therapy.
To determine the bioavailability of C1632 in rats, UHPLC-MS/ MS was applied. C1632 was delivered orally and intravenously, and blood samples were collected from the tail at indicated time points.
As shown in Figure 1C, there was a second peak at 1 h in the oral administration group. So, C1632 might be distributed to the liver and intestines. The mean concentration-time curves are shown in Figure 3C,D. The pharmacokinetic parameters were calculated and are presented in Table S1. The bioavailability of C1632 is 44.45%.
In addition, a cocktail assay was performed to evaluate the impact of C1632 on CYP450s. As shown in Figure 1E, the activities of four primary CYP450 isoenzymes, CYP3A2, CYP2C11, CYP2D1, and CYP2B1, were measured in the presence or absence of C1632.
Although the results indicated that C1632 could significantly inhibit their activity, the concentration of C1632 was relatively high.
Therefore, the effect of C1632 on liver drug enzymes is limited.

| C1632 suppresses the expression of LIN28 and blocks FGFR1-mediated signalling in NSCLC A549 and A549R cells
On the basis of data from the TCGA database, 43 we validated that both adenocarcinoma and squamous cell carcinoma of lung patients, which comprise 40% and 25% of NSCLC, 1 respectively, express much higher LIN28 mRNA levels in tumour tissue than in normal tissue (Figure 2A,B, Figure S2). Meanwhile, survival analysis showed that the patients expressing higher LIN28 showed lower survival rates compared with patients with low LIN28 expression ( Figure 2C). Additionally, blockage of the FGFR1 signalling pathway has been verified as a practical therapeutic strategy in NSCLC. 7,9,[26][27][28]30 Moreover, the positive correlation between LIN28B and FGFR1 had further been verified by LIN28B knocked down and FGFR1 inhibited experiments ( Figures S3 and S4).
As LIN28 and FGFR1 are strongly related to the progression and prognosis of lung adenocarcinoma patients, and resistance to cisplatin is a major obstacle for the success of NSCLC therapy, 44 we selected the adenocarcinoma cell line A549 and the cisplatin-resistant cell line A549R to further validate the inhibitory effects of C1632 on LIN28 and FGFR1. As expected, the qPCR results showed that C1632 decreased LIN28 mRNA levels in both A549 and A549R cells, especially in A549R cells ( Figure 2D,E). Our western blot assay showed that C1632 inhibited the expression of FGFR1 as well as FGFR1 phosphorylation and the phosphorylation of the downstream kinase MAPK, 45 which was abnormally expressed in a series of cancers, and is involved in the regulation of cell proliferation, migration, survival, and apoptosis in a dose-dependent manner ( Figure 2F,G, Figure S5). The fact that C1632 suppresses the expression of LIN28 and blocks FGFR1 signalling supports the idea that it has the potential to serve as an anti-NSCLC drug ( Figure 2D-G, Figures S3 and S4).

| C1632 inhibits the migration of A549 and A549R cells by decreasing the phosphorylation of focal adhesion kinase and the expression of MMP-9
The increase in LIN28 or FGFR1 levels in NSCLC promotes cell migration and proliferation. 23,24,27 Therefore, we investigated the effects of C1632 on A549 and A549R cell migration. A cell adhesion assay that determines the adhesion between cells and matrix was performed in A549 and A549R cells. As shown in Figure 3A,B, both cell lines displayed a significant decrease in cell adhesion to the matrix after treatment with C1632 in a concentration-dependent manner.
Cell adhesion is often associated with cell migration. 42 Therefore, we performed a scratch-wound healing assay to determine the migration rate of A549 and A549R cells in the presence or absence of C1632. Our results showed that C1632 significantly decreased cancer cell migration ( Figure 3C and Figure S6A). After 72 h of treatment, in the control A549R group, 97% of the scratch was covered by migrated A549R cells, while only 60% of the scratch was covered when cells were treated with 60 mg/L C1632 ( Figure 3D). Similarly, C1632-treated A549 cells (60 mg/L) showed a significant decrease in migration rate ( Figure S6B). These results demonstrate that C1632 inhibits the migration of cancer cells.
To further confirm the reduced migration ability of cancer cells after C1632 treatment, a Transwell migration and invasion assay was performed ( Figure 3E and Figure S6C). The quantitative data demonstrated that A549R and A549 cells treated with C1632 exhibited decreased migration compared with untreated cells ( Figure 3F and Figure S6D).
Furthermore, we found that C1632 suppressed the phosphorylation of focal adhesion kinase (FAK) and the expression of matrix metalloproteinase-9 (MMP-9; Figure 4 and Figure S7), which have been widely implicated in the adhesion, invasion, and migration of cancer cells. 46 First, IF results showed that C1632 reduced the distribution and the foci formation of FAK in A549R cells in a concentrationdependent manner ( Figure 4A,B, and Figure S7). In addition, western blot results further showed that C1632 efficiently decreased the phosphorylation of FAK and the expression of MMP-9 ( Figure 4C,D).
Taken together, these findings support the conclusion that C1632 . € C1632 inhibits the enzymatic activity of various CYP450s. The microsomal incubation assay was employed to study the inhibitory effects of C1632 on CYP3A2, CYP2B1, CYP2C11, and CYP2D1 of the rat. The probe substrates were added into the system, and their metabolites were detected. Relative enzymatic activity was calculated and plotted. Values are the average ± SD of three independent experiments. p values were calculated using the unpaired Student's t-test (*p < 0.05, **p < 0.01, ***p < 0.001)

| C1632 suppresses the colony formation of A549 and A549R cells by inhibiting DNA replication and inducing G0/G1 cell cycle arrest
C1632 decreased the viability of A549 and A549R cells at a high dose (60 mg/L), but not at a low dose (15 mg/L; Figure 5A,B). It is worthy to note that a high dose (60 mg/L) of C1632 almost had no cytotoxicity on the human normal foetal lung fibroblast cell line MRC5 ( Figure 5C). Consistent with the MTT assay, the colony formation assay showed that C1632 inhibited the formation of colony units in a dose-dependent manner, and the inhibition of colony formation was stronger in A549R cells than in A549 cells, in agreement with the cytotoxicity on A549 and A549R cells ( Figure 5D,F). The annexin V/PI apoptotic assay and the SAβ gal staining assay revealed that C1632 treatment, even at a high dose, did not significantly increase apoptosis or senescence ( Figures S8 and S9), suggesting that the inhibitory effects of C1632 on colony formation are not due to cytotoxicity-induced cell death.
Additionally, an Edu staining assay and flow cytometry were performed to further investigate whether C1632 inhibited the colony formation of A549 and/or A549R cells by DNA replication inhibition and cell cycle arrest. The results showed that C1632 treatment led to a significant inhibition of DNA replication in A549 and A549R cells in a dose-dependent manner ( Figure 6A

| C1632 suppresses the growth of A549R xenograft tumours in mice
The above results prompted us to examine the endogenous antitumour activity of C1632 on A549R xenograft tumours in mice.
Two weeks after injection with the cancer cell inoculum, and then every 2 days thereafter, mice were injected in the caudal vein with 30 mg/kg C1632. Although tumours were still visible after 18 days in the treated group, the tumour size was smaller than in the untreated group (untreated, mean ± SD = 2.35 ± 0.43 g; treated, mean ± SD = 1.36 ± 0.27 g; p < 0.05) ( Figure 7A,B). In  Figure 7C). Treatment did not affect the body weight of mice inoculated with A549R cells ( Figure 7D).
These results indicate that C1632 inhibits the growth of A549R xenograft tumours in mice and had no the toxi-side effects on the body.

| DISCUSS ION
It is now recognized that tumour drug distribution and bioavailability are important factors for effective tumour treatment. 47,48 Our results demonstrated that C1632 mainly accumulated in the lung after oral administration, with up to 44.45% bioavailability and limited Previous studies showed that either LIN28 or FGFR1 is strongly correlated with the progression of NSCLC, 22,27 and FGFR1 inhibitors achieved a definite therapeutic effect of NSCLC in the clinic and in an animal model. 30,31,34 However, FGFR1-targeted therapies are susceptible to drug resistance, 1,26,28 and there is still no LIN28 inhibitor available for NSCLC treatment. In this study, we demonstrated that it had a positive correlation between FGFR1 and LIN28B ( Figures S3 and S4), and C1632 suppressed the expression of LIN28 and blocked FGFR1 signalling in NSCLC A549 and A549R cells (Figure 2), resulting in an inhibition of migration (Figure 3 and Additionally, LIN28 has also been reported to promote cancer cell metastasis. 19 Thus, the anti-migration effects of C1632 (Figure 3 and Our results also reveal that C1632 treatment inhibited DNA replication of NSCLC A549 and A549R cells and induced G0/G1 cell cycle arrest ( Figure 6), indicating that the anti-NSCLC effect of C1632 is not only due to increased cell death (Figures S8 and S9). Interestingly, compared with A549 cells, C1632 exerts the same or even stronger anti-migration and anti-proliferation effects on A549R cells, regardless of drug resistance (Figures 3 and 4 and Figure S6).
Collectively, these results revealed that C1632 simultaneously suppressed LIN28 expression and blocked FGFR1 signalling and that C1632 is able to rapidly inhibit migration and proliferation of NSCLC cells, regardless of drug resistance, in vivo, indicating that it has the potential to act as an anticancer agent for NSCLC treatment.

ACK N OWLED G EM ENTS
The present study was funded by National Science Foundation of

CO N FLI C T O F I NTE R E S T S
The authors confirm that there are no conflicts of interest. Writing -review & editing (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
All data in this study are available if requested.